Oncotype DX maker Genomic Health announced last week that two studies recently presented at an international oncology conference could lead to the development of new predictive tests to help determine which patients are likely to respond to the targeted colon cancer drug Erbitux and the breast cancer chemotherapy Taxotere.
The first study, by Baker et al., debuts the initial results of Genomic Health's collaboration with Bristol-Myers Squibb and Imclone Systems to discover and validate markers associated with disease control and improved progression-free survival time in colon cancer patients treated with Erbitux, commonly known as cetuximab.
The second study, by Goldstein et al., used immunohistochemistry and quantitative RNA expression measured by RT-PCR to evaluate the predictive utility of progesterone receptor protein expression for 371 genes, including the current Oncotype DX 21-gene panel, for treatment of breast cancer either with doxorubicin plus cyclophosphamide, or with doxorubicin plus Taxotere, also known as docetaxel. This study was led by researchers from the Eastern Cooperative Oncology Group.
Both studies were presented at the American Society of Clinical Oncology’s annual meeting earlier this month in Chicago.
"These new data represent encouraging results of our work with leading cancer cooperative groups and drug manufacturers to discover and develop genomic tests to determine which patients are likely to benefit from treatment with targeted therapies," Steven Shak, Genomic Health’s chief medical officer, said in a statement.
Joffre Baker, Genomic Health’s chief scientific officer, told Pharmacogenomics Reporter this week that any tests developed from these collaborations will have a technology base similar to its flagship product, Oncotype DX.
Oncotype DX is the only FDA-approved genomic test that predicts the likelihood of chemotherapy benefit as well as the likelihood of disease recurrence for early-stage breast cancer patients. The test uses RT-PCR to perform a quantitative RNA analysis of fixed paraffin-embedded tumor tissue.
Genomic Health officials have previously stated that the company intends to expand the indication for Oncotype DX into other cancer types. And while the test previously yielded just a recurrence score, in the quarter ended March 31 the company began reporting measures of quantitative gene expression for estrogen and progesterone receptors [see PGx Reporter 01-16-2008]. This additional data, Genomic Health hopes, will improve physicians’ ability to determine the appropriate treatment for breast cancer patients.
In his statement, Shak also underscored the importance of collaborating with drug and diagnostic companies to develop Rx/Dx combination products.
"We believe this is an important step forward in demonstrating the potential of a diagnostic and therapeutic partnership in advancing the field of personalized medicine,” he said.
Pinpointing Erbitux Response
In Baker et al., researchers looked for KRAS gene mutations in formalin-fixed, paraffin-embedded tumor samples from 226 colon cancer patients archived from three Erbitux studies, and also for the expression of 102 previously identified candidate genes that may be associated with drug-related disease control and progression-free survival.
"Based on these results, we believe there is a potential to develop a multi-gene test comprising KRAS mutation status, in combination with the expression levels of a small number of genes to select patients for cetuximab.”
Findings from the retrospective study suggest that a “multi-parameter marker test comprising KRAS mutation status in combination with the expression levels of a small number of genes could be developed further to select patients likely to respond to Erbitux.
“Applicability of this test to cetuximab combination therapy should be explored,” the study authors concluded.
Out of 226 analyzed samples, 82 patients, or 36 percent, had KRAS mutations and a significantly lower disease control rate than those with the wild-type version of the gene. Disease control, a measure of clinical benefit, is defined as the frequency of both tumor shrinkage and disease stability.
Additionally, quantitative expression of 40 genes was “significantly associated” with disease control, researchers found.
"Based on these results, we believe there is a potential to develop a multi-gene test comprising KRAS mutation status in combination with the expression levels of a small number of genes to select patients for cetuximab," Baker, the lead author of the study, said in a statement.
Shak told Pharmacogenomics Reporter that Genomic Health is discussing additional studies with Bristol-Myers Squibb and Imclone to further define the number of candidate genes that will be included in any KRAS mutation test that is developed out of the collaboration.
In Goldstein et al., researchers from ECOG used IHC and quantitative RNA expression measured by RT-PCR to evaluate the predictive utility of progesterone receptor protein expression of 371 genes, which included those in the Oncotype DX 21-gene panel, for treatment either with doxorubicin plus cyclophosphamide, or with doxorubicin plus Taxotere.
Samples were collected from 734 patients who received at least three to four treatment cycles of either treatment regimen.
In the study, researchers found that progesterone receptor protein expression may be a marker for differential taxane benefit, but that the finding will require validation in other datasets.
“A genomic classifier predicting differential benefit was identified in HR-positive tumors with a recurrence score greater than 18, and if externally validated, might be useful in defining differential benefit in patients with HR-positive disease,” the study authors concluded.
According to Genomic Health, the study identified a number of candidate genes, expression of which predicted docetaxel response for patients with hormone receptor positive disease who had an Oncotype DX recurrence score greater than 18, and thus were classified to have an intermediate risk of cancer recurrence.
Furthermore, researchers found conflicting predictions when samples were analyzed by IHC versus RT-PCR. “There was a weak benefit for docetaxel in progesterone receptor-negative and for cyclophosphamide in progesterone receptor-positive disease by central IHC but not when genomic progesterone receptor was evaluated by RT-PCR,” the study authors found.
Meanwhile, in the study, Oncotype DX RS did not predict differential benefit when patients were treated with docetaxel versus cyclophosphamide in hormone-receptor positive disease.
In Goldstein et al., “a genomic classifier predicting differential benefit was identified and, if validated through additional studies, might be useful in defining differential benefit of docetaxel,” Genomic Health said in a statement.
According to Baker, in the study, more than 80 genes were identified as candidate predictors of taxane response. The company is planning additional studies to narrow and validate the most predictive genes to develop into a new panel to gauge treatment benefit with docetaxel.
The company said that it is not providing any additional details regarding this study, or its diagnostic plans for this indication.